R walkthrough

This walkthrough runs the everyday loop from R: initialize a project, add a file, check its status, delete it, then get it back. It is the "I re-cloned a project and need its data" sequence. The CLI walkthrough covers the same steps from the terminal.

🔗Setup

options(width = 1000)
library(dvs)
library(here)

here() starts at /Users/elea/Documents/a2ai_github/dvs2-demo-repo

Create a project directory and a sibling storage directory, then save R's built-in Theoph dataset as a CSV.

storage     <- basename(tempfile(fileext = "_storage"))
new_project <- basename(tempfile(fileext = "_project"))
dir.create(here::here(storage))
dir.create(here::here(new_project))
dir.create(here::here(new_project, "data"))
write.csv(Theoph, here::here(new_project, "data/theoph.csv"), row.names = FALSE)

🔗1. Initialize

Point storage at the sibling directory. See dvs_init() for every parameter.

setwd(here::here(new_project))
dvs_init(here::here(storage))

DVS Initialized

🔗2. Add

Track the CSV with a message. See dvs_add().

setwd(here::here(new_project))
dvs_add("data/theoph.csv", message = "initial Theoph data")

# A tibble: 1 × 5
  path            outcome hash                                                                size stored_size
  <chr>           <chr>   <chr>                                                            <bytes>     <bytes>
1 data/theoph.csv copied  cdd978e51298006701f7b285aaf979933f0af6b179bbdf3347014af3bcd48c06  2.9 KB       842 B

🔗3. Status

The file is tracked and current. See dvs_status().

setwd(here::here(new_project))
dvs_status()

# A tibble: 1 × 8
  path            status  hash                                                                size created_by compression message             add_time           
  <chr>           <chr>   <chr>                                                            <bytes> <chr>      <chr>       <chr>               <dttm>             
1 data/theoph.csv current cdd978e51298006701f7b285aaf979933f0af6b179bbdf3347014af3bcd48c06  2.9 KB elea       zstd        initial Theoph data 2026-06-16 16:46:45

🔗4. Remove the file

Delete the local copy, the moment after a fresh clone where the data is not on disk yet.

unlink(here::here(new_project, "data/theoph.csv"))

🔗5. Status again

The file is now absent: tracked, but missing locally.

setwd(here::here(new_project))
dvs_status()

# A tibble: 1 × 8
  path            status hash                                                                size created_by compression message             add_time           
  <chr>           <chr>  <chr>                                                            <bytes> <chr>      <chr>       <chr>               <dttm>             
1 data/theoph.csv absent cdd978e51298006701f7b285aaf979933f0af6b179bbdf3347014af3bcd48c06  2.9 KB elea       zstd        initial Theoph data 2026-06-16 16:46:45

dvs reports three states: current, absent, and unsynced (on disk but not matching the stored hash). See Storage and meta files.

🔗6. Get

Restore the file from storage. See dvs_get().

setwd(here::here(new_project))
dvs_get("data/theoph.csv")

# A tibble: 1 × 3
  path            outcome    size
  <chr>           <chr>   <bytes>
1 data/theoph.csv copied   2.9 KB

🔗7. Status again

The file is current again. The loop is complete.

setwd(here::here(new_project))
dvs_status()

# A tibble: 1 × 8
  path            status  hash                                                                size created_by compression message             add_time           
  <chr>           <chr>   <chr>                                                            <bytes> <chr>      <chr>       <chr>               <dttm>             
1 data/theoph.csv current cdd978e51298006701f7b285aaf979933f0af6b179bbdf3347014af3bcd48c06  2.9 KB elea       zstd        initial Theoph data 2026-06-16 16:46:45

🔗Cleanup

unlink(here::here(new_project), recursive = TRUE)
unlink(here::here(storage),     recursive = TRUE)

🔗Next steps

• The R Package reference covers every parameter of dvs_init(), dvs_add(), dvs_status(), and dvs_get().
• The same loop from the terminal: CLI walkthrough.
• How dvs stores data: Storage and meta files.